digitalML Blog: The Key to the Intelligent Enterprise

  • API Strategy
  • Banking and Payments
  • Innovation
  • IT Modernization
  • TechTrends
  • Generic selectors
    Exact matches only
    Search in title
    Search in content
    Search in posts
    Search in pages

    What is an Abstraction Layer? What Makes it Different?

    July 31, 2018

      |   Reading Time: 3 minutes

    All large enterprises share one major problem – siloed systems. Yet, the information and services locked away in these silos is essential for moving forward with new digital initiatives. One of the reasons a company like Amazon can create such fear at even a hint of moving into a new market, is that they’ve evolved with modular IT systems which can be easily reconfigured to support new digital business offerings.

    So how do the established players tackle this issue? For years they’ve poured vast amounts of money, time and effort into integration technologies, but all this does is add more layers of technical debt and compounds the bigger problem.

    The idea of an abstraction layer has been around for a long time. Past efforts have never been able to deliver – either because initiatives have been focused at a very technical level which isn’t understood by the business, or because business teams have built an abstracted view of their capabilities which have little or no connection to the enterprise’s IT systems.

    A blended approach to an abstraction layer is key

    Thankfully, it’s now possible to take a blended approach to this abstraction layer – where it is modularized and organized in a way which can be understood by business teams, while still containing sufficient levels of detail for the technical audience to drill down to.

    This blended approach avoids communication mismatches and aligns focus to the real aims of the business. Agile principles ensure that both business and technical focuses are considered in the development stage, which in turn leads to software that better supports the overall business goals.

    A blended abstraction layer ensures these benefits are applied to technical capabilities, grouping them in modules that support business capabilities.

    An abstraction layer is surprisingly easy to build

    The best part about it? It’s surprisingly easy to build this abstraction layer, because most of the information needed for it already exists! This information can either be found within the enterprise already or easily copied across verticals.

    Tie services to the business functions they support in an abstraction layer

    Tie services to the business functions they support in an abstraction layer

     

    Auto-generate code directly from the abstraction layer

    Many of the software industry’s recent moves have generated a wealth of tools to allow rapid and scalable ways to include code. These include:

    • Cloud-based infrastructure and processes
    • Harnessing the power of containerization
    • Continuous development and deployment
    • Provisioning capacity on demand

    These moves have also enabled the possibility of code to be directly auto-generated from the abstraction layer. This subsequently means ideas can be quickly and cheaply turned into prototypes, then rapidly market tested and scaled out if the feedback loop is favorable.

    Abstraction enables IT modernization

    Having a blended abstraction layer not only supports digital initiatives – it is also the key to cleaning up all that technical debt. It enables modernization of siloed applications and repurposes them directly into the cloud.

    Repurpose existing functionality into reusable business capability building blocks in the abstraction layer

    Repurpose existing functionality into reusable business capability building blocks in the abstraction layer



    To learn more about how the ignite platform from digitalML makes it easy to build and manage your abstraction layer of APIs and Services, visit our platform page.

     

    About the Author
    Jeremy SindallJeremy Sindall
    Jeremy is CEO of digitalML. Having founded the company in 2000, he specializes in helping the world's largest companies reinvent themselves into Intelligent Enterprises through abstracted API and service management.